AbstractsMathematics

Correlation analyses of the seasonal changes and qualityof life with the ciinical presentations of otitis media in children3-8 years of age

by Snežana D Andrić-Filipović




Institution: Univerzitet u Beogradu
Department:
Year: 2016
Keywords: Запаљење средњег ува; сезона; мере слушања; упитници; развојни исход; скрининг деце са запаљенским променама средњегува
Posted: 02/05/2017
Record ID: 2105248
Full text PDF: https://fedorabg.bg.ac.rs/fedora/get/o:12512/bdef:Content/get


Abstract

Medicina / Medicine Otitis media (OM), in both the acute (RAOM) and chronic (OME) forms affecting young children, plus the developmental consequences, are a large burden on families’ well-being and on healthcare systems. The burden comes from the large numbers with these very common conditions, rather than extreme individual cases. Past research on OM has mostly not well addressed the need for precise assessment of severity and statistical handling of case flow, via optimum referral and treatment decisions. The OM8-30 (with its short form OMQ-14) questionnaire is being standardised for Europe through the international clinical study Eurotitis- 2. I contributed nearly one quarter of the cases (22.6%) and was regional coordinator for about a further 8.8%, making an overall contribution of one third of the cases (31.4%), so was given access to the whole database under agreements in place, for answering my research questions. The first study moved on from the known seasonality of incidence of respiratory infection plus OM and used the severity measures from OM8-30 on different aspects of OM presentation. It asked: in cases all with OM diagnosis, do the measured severities of the several disease facets vary systematically through the year, after the typical autumn respiratory virus spread triggers an annual cycle? And do the timings of maximum severity in the single-consultation crosssectional data reflect the cause-related sequence from upstream respiratory infection, through ear infection score (ESS) and measured and reported hearing (HL and RHD) to downstream developmental outcomes? To answer this question with maximum precision, I worked with the study statistician to develop a method for locating annual peak severity which is labourintensive but robust: fitting a series of 27 lagged sinusoid functions I also developed a control framework fitting, as possible additive confounders, the background determinants of severity (particular centre, age, socioeconomic status (SES), length of history and particular diagnosis; sex was usually not significant). The novel results with these methods were clear for the upstream stages. The cross-sectional data show distinct annual cycles of case severity, with the ordering URTI(ESS, HL)RHD, and even close correspondence of estimated absolute delays with those in other true time-series (longitudinal) data. Season (or date) thus has to join the list of adjuster variables in the control framework underpinning standardisation... Advisors/Committee Members: Milovanović, Jovica, 1968-.